How to write a dissertation abstract (with examples)

The six elements every abstract needs

An abstract is not an introduction. An introduction sets up a question; an abstract answers it. Someone who reads only your abstract should be able to state your finding.

Allocate roughly one to two sentences per element, which lands you close to 250 words naturally.

• Problem — the gap or issue, in one sentence.
• Aim — what your study set out to establish.
• Method — design, sample and analysis, compressed.
• Findings — the actual result, with numbers or named themes. This is the part people omit.
• Implications — what it means for practice, policy or theory.
• Limitation — the honest caveat, stated in one clause.

How long should a dissertation abstract be?

Most UK and Australian institutions specify 250–300 words for a master's dissertation and 300–350 for a doctoral thesis. Undergraduate dissertations are often shorter at 150–250 words.

Check your handbook — this is one of the few places where an over-length submission is caught automatically, because abstracts are frequently submitted through a separate form field with a hard character limit.

Worked example — social sciences

Widening participation policy has increased university entry among first-generation students, but retention gaps persist and the mechanisms behind them remain poorly understood. This study examined how institutional belonging predicts first-year continuation among first-generation undergraduates at a post-1992 UK university. A mixed-methods design combined a survey of 412 first-year students with 18 semi-structured interviews, analysed using hierarchical regression and reflexive thematic analysis. Belonging predicted continuation intention more strongly than either prior attainment or financial stress (β = .41, p < .001), and interviews identified three mechanisms: recognition by academic staff, peer-group formation in the first six weeks, and the legibility of institutional processes. These findings suggest retention interventions targeting early social integration may outperform those targeting academic support alone. The single-institution design limits generalisability to comparable post-1992 providers.

Note what this does: it reports a coefficient, names the three themes, and states an implication. It never says 'this study will explore'.

Worked example — business and management

Small and medium enterprises adopted remote working rapidly during 2020–21, but the productivity consequences of retaining hybrid arrangements are contested. This study assessed the relationship between hybrid working intensity and team productivity in 63 UK SMEs. Panel data covering 24 months were analysed using fixed-effects regression with firm-level controls. Productivity rose modestly at two to three remote days per week (+4.2%) but declined beyond four days (−6.8%), indicating a non-linear relationship rather than the linear effect assumed in much of the existing literature. Managers should therefore treat hybrid intensity as an optimisation problem rather than a binary choice. Self-reported productivity measures introduce potential common-method bias.

The five mistakes that cost marks

These recur in almost every batch of dissertations we review.

• Promising instead of reporting — 'this study investigates' with no result stated.
• Citations in the abstract. An abstract stands alone; it does not cite.
• Undefined abbreviations. Spell out everything on first use, even if defined later.
• Background bloat — three sentences of context leaves no room for findings.
• Writing it first. Write the abstract last, when you know what you found.